Re: Discrete Chaos and Learning Orgs LO869

Fred Reed (freed@cc.atinc.com)
Fri, 21 Apr 95 13:12:29 EST

Reply to LO 852 (I think)

Mike, I don't have a Kampis in front of me, nor am I a Systems Dynamics
guru, but I do have some experience in modelling of complex man-machine
systems. The one thing I do remember coming away from Kampis with is a
sinking feeling about the legitimacy of many such models, and
particularly the role they play in making critical real-life decisions.
When one is dealing with component-systems, or of systems containing
component systems, it is entirely possible to be *completely* wrong with
a formal model of any size/complexity/cost for the reason you stated.

When such models are used for exploration and furthering understanding,
they are quite useful in spite of this limitation. *However*, when we
start making decisions (from how to actually run an organization, to
where to put a radioactive waste dump) based on systems dynamics models
we will (and have!) caused serious problems.

If you remember, this was the distinction of "complexity" I was trying
to make some time ago, in that even chaos/non-linear dynamics cannot
cover the complexity of such component systems.

An interesting paper on the risks of modelling is "Information
Technology and Mathematical Modelling, The Software Crisis, Risk and
Education Consequences" by Bernhelm Booss-Bavnbek in Computers &
Society, Vol 19, No. 3 - Sep 1989.

Fred Reed
freed@cc.atinc.com
_______________________________________________________________________________
Subject: Re: Discrete Chaos and Learning Orgs LO852
From: learning-org@world.std.com at ~INTERNET
Date: 4/21/95 12:19 PM

Replying to LO826 --

<snip>
I also have a question for those who have put in their time with systems
dynamics. Does the operational definition of a "component system" apply
to the production problems mentioned? If so, then can System Dynamics
provide anything useful given that "in a component-system we cann know but
a fraction of all components, and this fraction does not represent well
the whole, since the components are all so different"? This statement
implies that "Component-systems produce new elements but dynamics can only
be written down for the known elements." (Quote from 'Self-Modifying
Systems in Biology and Cognitive Science by G. Kampis, Pergamon Press)

Thanks for the stimulating material,

-- 
Mike McMaster      <Michael@kbddean.demon.co.uk>
    "Intelligence is an underlying organisational principle
     of the universe.  The 'logos principle' is hidden and
     perceptible only to the intelligence."   Heraclitus